# US 2023/0266879 A1 · System, method and computer-readable storage medium for direct memory accesses

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2023/0266879 A1
- Document ID: US20230266879A1
- Status: Published application
- Kind: A1
- Application number: 18/162,836
- Filed: 2023-02-01
- Published: 2023-08-24
- Earliest claimed: 2022-02-18
- Assignees: Samsung Electronics Co., Ltd.
- Applicants: Samsung Electronics Co., Ltd.
- Inventors: Younghwan JOO; Venkata Rama Krishna MeKa
- Web page: https://patentgrep.com/patent/US20230266879A1

## CPC

- G06F 3/061: {Improving I/O performance}
- G06F 3/0656: {Data buffering arrangements}
- G06F 3/0659: {Command handling arrangements, e.g. command buffers, queues, command scheduling}
- G06F 3/067: {Distributed or networked storage systems, e.g. storage area networks [SAN], network attached storage [NAS]}
- G06F 3/0679: {Non-volatile semiconductor memory device, e.g. flash memory, one time programmable memory [OTP]}
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)

## Abstract

Provided is a system including a memory, a device including a plurality of direct memory accesses (DMAs), each of the plurality of DMAs configured to read from or write to the memory, and at least one processor configured to control the device by executing a program, wherein executing the program includes implementing a kernel configured to abstract the plurality of DMAs into a node, and an application configured to be executed on the kernel and access the device through the node.

## Claims

### Claim 1

A system comprising:
a memory;
a device comprising a plurality of direct memory accesses (DMAs), each of the plurality of DMAs configured to read from or write to the memory; and
at least one processor configured to control the device by executing a program, wherein executing the program includes implementing
a kernel configured to abstract the plurality of DMAs into a node; and
an application configured to be executed on the kernel and access the device through the node.

### Claim 2 (depends on 1)

The system of claim 1, wherein the node comprises information regarding the plurality of DMAs.

### Claim 3 (depends on 2)

The system of claim 2, wherein the node comprises
a first frame accessed by the application,
a plurality of second frames respectively corresponding to the plurality of DMAs, and
a plurality of third frames comprising information regarding buffers accessed by the plurality of DMAs, respectively.

### Claim 4 (depends on 3)

The system of claim 3, wherein
the first frame comprises a plurality of pointers respectively corresponding to the plurality of second frames, and
the plurality of second frames each comprise at least one pointer corresponding to at least one of the plurality of third frames.

### Claim 5 (depends on 3)

The system of claim 3, wherein the first frame comprises setting information regarding a result frame output by an operation of the device.

### Claim 6 (depends on 3)

The system of claim 3, wherein the plurality of second frames comprise information regarding addresses and sizes of the plurality of DMAs or ID information of hardware, the ID information of the hardware including ID information of the plurality of DMAs.

### Claim 7 (depends on 3)

The system of claim 3, wherein the plurality of third frames comprise an address of a buffer accessed for operations of the plurality of DMAs.

### Claim 8 (depends on 1)

The system of claim 1, wherein the device is configured to process image data stored in the memory under control by the at least one processor.

### Claim 9

A method performed by at least one processor by executing a program, the method comprising:
providing, by a kernel implemented as part of executing the program, a node abstracting a plurality of direct memory accesses (DMAs), each of the plurality of DMAs configured to read from or write to a memory included in a device; and
accessing, by an application implemented as part of executing the program, the device via the node.

### Claim 10 (depends on 9)

The method of claim 9, wherein the node comprises information regarding the plurality of DMAs.

### Claim 11 (depends on 10)

The method of claim 10, wherein the node comprises
a first frame accessed by the application,
a plurality of second frames respectively corresponding to the plurality of DMAs, and
a plurality of third frames comprising information regarding buffers accessed by the plurality of DMAs, respectively.

### Claim 12 (depends on 11)

The method of claim 11, wherein the first frame comprises a plurality of pointers respectively corresponding to the plurality of second frames, and
the plurality of second frames each comprise at least one pointer corresponding to at least one of the plurality of third frames.

### Claim 13 (depends on 11)

The method of claim 11, wherein the first frame comprises setting information regarding a result frame output by an operation of the device.

### Claim 14 (depends on 11)

The method of claim 11, wherein the plurality of second frames comprise information regarding addresses and sizes of the plurality of DMAs or ID information of hardware, the ID information of the hardware including ID information of the plurality of DMAs.

### Claim 15 (depends on 11)

The method of claim 11, wherein the plurality of third frames comprise an address of a buffer accessed for operations of the plurality of DMAs.

### Claim 16 (depends on 9)

The method of claim 9, wherein the device is configured to process image data stored in the memory under control by the at least one processor.

### Claim 17

A non-transitory computer readable recording medium having stored thereon a program to be executed by at least one processor, wherein, when the program is executed by at least one processor, the program performs a method in which the at least one processor controls a device comprising a plurality of direct memory accesses (DMAs), each of the plurality of DMAs configured to read from or write to a memory, and the method of controlling the device comprising:
providing, by a kernel implemented as part of executing the program, a node in which the plurality of DMAs are abstracted; and
accessing, by an application implemented as part of executing the program, the device through the node.

### Claim 18 (depends on 17)

The non-transitory computer readable recording medium of claim 17, wherein the node comprises information regarding the plurality of DMAs.

### Claim 19 (depends on 18)

The non-transitory computer readable recording medium of claim 18, wherein the node comprises
a first frame accessed by the application,
a plurality of second frames respectively corresponding to the plurality of DMAs, and
a plurality of third frames comprising information regarding buffers accessed by the plurality of DMAs, respectively.

### Claim 20 (depends on 19)

The non-transitory computer readable recording medium of claim 19, wherein the first frame comprises a plurality of pointers respectively corresponding to the plurality of second frames, and
the plurality of second frames each comprise at least one pointer corresponding to at least one of the plurality of third frames.
